EViews.com

Hello,

I am no EVIEWS expert and I need some help with an estimation of a complex GJR-GARCH model with a dummy-variable in the variance equation (See model below).
The estimation should use robust standard errors (Bollerslev/Wooldbridge, 1992). I have no problems with the mean equation, but I am not able to create the multiplicative dummy[y(D) D(t)] in the variance equation. Is it possible to create such a multiplicative dummy in EVIEWS? Can I write a programm in EVIEWS that can solve my problem? I would be thankful for some advices or source codes that I could modify. Thanks in advance.
W = Dummy for Days of the week
r = index returns
r (US/HS) = index returns of foreign stock markets
I = dummy-variable that takes the value of one if the past innovations are negativ (asymetric term of the GJR-GARCH)
For a similiar model see Martin T. Bohl, Michael Schuppli and Pierre L. Siklos (2009): Stock return seasonalities and investor structure: Evidence from China's B-share markets

[img][IMG=http://img811.imageshack.us/img811/7152/gjrgarch.th.png][/img]
[/img]

Using the built in GARCH estimation routine, no. But you could create a LogL object with the correct likelihood specified. See the example logl programs in the manual for details.